The Deep Clean

If your organization has heeded the advice of conference speakers everywhere, it has recorded information about donors into a database, donor-management system or constituent-relationship management system. Fifty thousand records later, it has effectively developed a phone book. It’s a large database, but it might not be a deep one.

Here we examine three major stages to increasing and taking advantage of database depth.

Stage 1. Increase reachability
Core to every record is basic contact information, such as e-mail, phone and mailing address. These enable corresponding outreach campaigns in the most rudimentary sense.

Since, generally, people do not donate unless they are asked to, if they cannot be reached, then they will not donate. The metric to optimize here is “reachability,” namely, how many of the contacts can be reached right now. This number decreases over time not because the database is broken, but because the world is changing around it. People move, change jobs, get married, etc. This is called “data rot” and is one of the most common reducers of reachability.

Completeness means that every contact has every form of contact information, and correctness means that this information is actually usable. Some organizations set interns or administrative staff to the arduous task of scouring the Web to improve completeness and correctness. If that is your organization, stop!

Data-cleansing services exist to automate this process, enhancing data quality by enabling human resources to be applied to tasks that only humans can handle. We have seen up to a 900 percent increase in year-over-year donations as a result of such a service combined with more effective utilization of staff members.

Here’s what to look for in an effective data-cleansing service:

1. Missing contact information is appended: For example, if a record has a name and address, the service should be able to provide an e-mail address and phone number. Completeness of the incoming data set has a big effect on service quality because, after all, there are a lot of John Smiths.

2. All information is standardized: To avoid doubt and ensure reachability, phone numbers should all have area codes, ZIP codes should have +4s and e-mail addresses should conform to format standards. Databases also have an easier time counting, searching and correctly associating standardized data.

3. Contact information is validated: 555-555-5555 and fake@address.com may be correctly standardized yet still unusable. A good cleaning service runs checks just prior to delivering updated information to increase the odds that information it provides is currently usable and is correct for the contact in question.

4. Price and quality claims are transparent: Be suspicious of any service that requires a commitment to pay before it gives a price estimate. Some give wide ranges instead of analyzing the database up front. Expect to get a free analysis and some form of guarantee about what happens if any bad data is provided.

The savings in time, materials and stress, combined with the increase in the number of people that will respond to a campaign, can greatly outweigh the price of data-cleansing services. To test its customer service, ask the provider to help you calculate the return on investment from better data for your next campaign.

Stage 2. Add data types to enable donor analysis
Who donated how much and to which campaigns? That’s basic bookkeeping. What you don’t know about your constituents can inhibit charitable contributions. The deep dive to understand the characteristics of who gives and why requires more information about each individual. Collecting this information takes a lot of effort on both the part of the organization and the individual if traditional methods like questionnaires are utilized.

Retailers from your local grocer to Amazon have long faced the same challenges. The same solution they use is available to the nonprofit world: Utilize data services to append estimated demographic, psychographic, income, debt, interest and more to each donor in your database.

Some services can offer up to a thousand fields of information about consumers, all collected from public and permission-granted sources. This is not an article about privacy, so just know that retailers often sell the data you fill out on rebate offers and in service applications. Data services aggregate and cross-reference this information for accuracy.

By better understanding who might want to donate to your cause, you can save resources and respect privacy by discerning who not to ask for donations. Furthermore, it is courteous and useful to each recipient of your ask when you tailor your message to his or her individual interests and circumstances.

To take advantage of this form of database depth, be sure to record into the database as much as possible about what is learned from each donor from every interaction. The easiest way to learn a lot is to purchase additional fields of data. This costs pennies to tens of cents per record, depending on which kinds of data are sought and in what volume.

Don’t expect providers to reveal the exact sources of their data, but do expect them to be up-front with pricing and complete estimates. Moreover, look for providers that establish ongoing relationships with their clients by way of their quality guarantees, access to online platforms, and abundant but not unlimited free customer support.

Stage 3. Optimize campaigns using technology
With great care and effort, a human can inspect giant spreadsheets created from Stages 1 and 2 to build mental models of trends and individuals. Luckily, however, there are technologies and consultants that can simplify this task considerably.

To recap, in Stage 1, increased reachability means each contact’s information is complete and up to date. In Stage 2, this “good data” is matched against providers’ databases to yield additional information about individuals. Naturally, the better Step 1 goes, the better the results are in Stage 2. Similarly, the better Stage 2 goes, the more that algorithms in Stage 3 can do to tease out trends and assist with the detail-oriented work of either tailoring a message or determining not to bother that constituent.

Much of the technology available to achieve these ends requires a somewhat sophisticated understanding of statistics. The best bet is to find a company that can both provide up-to-date data and perform the analysis of it. Expect such companies to have privacy policies and to require that your organization confidentially share a significant portion of your database in order to get the job done.

There is a different mentality, reason and purpose behind each donor to each different organization. The only way to truly understand donors is to look at the details of their donation histories in combination with enough fields of data that one algorithm (or very smart person) could find a pattern within the Stage 2 data.

This donor analysis can distinguish those who are likely to donate from those who are not. As with the previous stages, look for companies that provide good customer service in the forms of being willing to carefully explain the analysis, load and perform a basic analyses of the data for free, and work with you to estimate and measure the return on investment from the Stage 3 outcomes.

Benjamin Ashpole is CEO of Updentity. Reach him at bashpole@bashpole.com

Brandy Coomer is chief interaction manager at Updentity. Reach her at brandy.coomer@updentity.com